Research On Intelligent Diagnosis Technology Of Mechanical Faults Based On Probability Graph Model

With the continuous development of modern industrial technology,the working environment of machinery and equipment is becoming more and more complex,and its components are prone to damage,and machine failure will damage production equipment,reduce product quality,and even cause casualties.In this paper,a mechanical fault intelligent diagnosis system based on probability graph model is studied,which is based on probability graph model technology,establishes a mechanical probability diagram fault diagnosis model,and outputs the probability value of equipment failure by calculating the symptom status value under real-time working conditions of equipment and entering the diagnostic model,and takes the fault type with the highest probability as the diagnosis result.The research content of this paper includes the structural design of probability graph fault diagnosis model,the prior probability setting method and the conditional probability setting method.Finally,the typical failure test of reciprocating compressor was simulated for the research object in the past,and the proposed method was verified and analyzed based on the test data,and the specific research content of this paper is as follows:(1)Research and establish the structure of mechanical fault intelligent diagnosis model based on probability map,design a probability map fault diagnosis model structure including three layers of prior probability layer,fault layer and symptom layer,combine typical mechanical faults and their observable signs,and analyze the calculation and reasoning steps of fault occurrence probability.(2)The prior probability setting method of the intelligent diagnosis model is studied,and the prior probability value represents the actual probability of mechanical failure.By introducing the time factor,a prior probability setting method based on the failure rate curve is proposed,and the prior probability value is determined by dividing different time intervals,and the set prior probability value is more consistent with the actual operation of the equipment,which is conducive to the subsequent accurate calculation of the probability value of each fault.(3)The conditional probability setting method of intelligent diagnosis model is studied,and a conditional probability setting method based on analytic hierarchy method to determine the symptom association level is proposed,and the strength of the association of signs to failure is accurately evaluated by establishing a symptom association level evaluation model,and according to the correspondence between the symptom association level and the conditional probability value,a complete probability map fault diagnosis model conditional probability table can finally be obtained.The proposed method gets rid of the dependence on sample size,and the conditional probability value setting is more scientifically accurate and convincing.(4)Applying analysis,in the past,the composite compressor established its probability map fault diagnosis model for the research object,and calculated the prior probability and conditional probability based on the methods described in this paper.In addition,the typical fault simulation test of reciprocating compressor is carried out,and the fault diagnosis test is carried out according to the test data,which verifies the effectiveness of the proposed method.